Supporting Online Material for

Transcription

1 Supporting Online Material for The Competitive Cost of Antibiotic Resistance in Mycobacterium tuberculosis Sebastien Gagneux,* Clara Davis Long, Peter M. Small, Tran Van, Gary K. Schoolnik, Brendan J. M. Bohannan *To whom correspondence should be addressed. This PDF file includes: Materials and Methods Tables S1 to S5 References Published 30 June, Science 312, 1944 (2006) DOI: /science

2 Supporting Online Material Material and Methods Isolation of spontaneous rifampin-resistant mutants Single colonies were isolated from the antibiotic-susceptible clinical strains CDC1551 and T85. The colonies were sub-cultured and aliquots frozen at 80 o C for later use. For both strains, four 1000ml roller bottles with 50ml Middlebrook 7H9 broth (Difco), supplemented with ADC (Difco) and 0.05% Tween were inoculated and grown to late log phase at 37 o C. From each roller bottle, 200µl of culture were inoculated onto each of four Middlebrook 7H11 (Difco) plates with 5µg/ml rifampin and incubated at 37 o C. All the rifampin-resistant colonies that appeared after 4 weeks of incubation were subcultured in Middlebrook 7H9 broth and heat-inactivated for direct sequencing of the rpob gene using the following PCR/sequencing primers: F-5 - TCGGCGAGCTGATCCAAAACCA-3, R-5 -ACGTCCATGTAGTCCACCTCAGA-3 ( A total of 52 colonies with the CDC1551 background were analyzed and 12 different rpob mutations identified. Eight of these have been reported in clinical settings (1). We included one representative of each of these mutations in our study. We also included one mutant which, to date, has not been described in clinical settings. We analyzed 63 colonies for T85, and identified seven distinct rpob mutations, six of which were also represented in the CDC1551 background. We included four T85 mutants in the study for which we had the corresponding fitness measurements in the CDC1551 background. 1

3 Clinical isolates Serial isolates were obtained from ten tuberculosis patients who developed resistance to rifampin during treatment. For each patient, one initial rifampin-susceptible and one subsequent rifampin-resistant isolate were genotyped by Spoligotyping according to standard procedures (2). For each isolate, single colonies were sub-cultured and aliquots prepared and stored at 80 o C for the competition experiments. The drug resistance profiles of all laboratory-derived and clinical isolates were confirmed by comparing their growth on Middlebrook 7H11 agar plates with no antibiotic, to their growth on plates supplemented with either rifampin (5 mg/l), isoniazid (0.2 mg/l), or streptomycin (2 mg/l). Competition assays A standard competition assay developed for E. coli by Lenski et al. (3) was adapted for M. tuberculosis as follows: In two 1000ml roller bottles, 100ml of Middlebrook 7H9 broth, supplemented with ADC (Difco), 0.2% Tween and 10g sterile 2mm diameter glass beads, were inoculated separately with an aliquot of the rifampin-susceptible and an aliquot of the rifampin-resistant strains. In this conditioning step, both strains were grown separately at 37 o C in a rolling incubator to stationary phase. Optical densities (OD 600 ) were measured daily until the OD 600 of both cultures stopped increasing. At stationary phase, 1.5ml of each culture was vortexed with sterile 0.3mm diameter glass beads for 10 seconds, serially diluted in Middlebrook 7H9 broth with 0.2% Tween, and equal amounts (~ 10 6 CFUs) co-inoculated into a new 1000ml roller bottle with 100ml Middlebrook broth, supplemented with ADC, 0.2% Tween and 10g sterile 2mm diameter glass beads. 2

4 For baseline colony-forming units (CFU) counts, the mixed culture was serially diluted and plated in triplicate on petri dishes with 25ml of Middlebrook 7H11 medium with and without 5µg/ml of rifampin. The competition cultures were incubated at 37 o C in a rolling incubator until reaching stationary phase. For endpoint CFU counts, 1.5ml of the mixed culture were vortexed as outlined above, serially diluted and plated in triplicate onto Middlebrook 7H11 plates with and without rifampin. Both baseline and endpoint plates were counted after four weeks of incubation at 37 o C. Statistical analysis For each competition experiment, the mean of the three plates was used for the calculation of the relative competitive fitness. At baseline and endpoint, the CFU counts on the rifampin plates indicated the number of rifampin-resistant cells in the mixed cultures. The number of susceptible cells was calculated by subtracting the number of resistant cells from the total cell number revealed by the CFU counts of the plain plates. The relative competitive fitness W of the rifampin-resistant compared to the rifampinsusceptible strain was calculated using the following formula (3): W= ln (R F /R I )/ln (S F /S I ) where R I and S I refer to the number of resistant and susceptible cells at baseline, respectively, and R F and S F to the number of resistant and susceptible cells at endpoint. Means of four to seven replicate competition assays were determined. The 95% confidence intervals were calculated based on the t-distribution with (N-1) degrees of freedom, were N refers to the number of replicates (3). Statistical test were performed with JMP (version 5.1, SAS Institute, USA). 3

5 Supporting Online References 1. M. O'Sullivan D, T. D. McHugh, S. H. Gillespie, J. Antimicrob. Chemother. (2005). 2. P. F. Barnes, M. D. Cave, N. Eng.l J. Med. 349, 1149 (2003). 3. R. S. Lenski, M. R. Rose, S. C. Simpson, S. C. Tadler, Am. Na.t 138, 1315 (1991). 4

6 Supporting Online Tables Table S1. Rifamin-resistant mutants of strain CDC1551 generated in vitro. Mutant rpob codon Wild type base pair Wild type amino acid Mutant base pair Mutant amino acid CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC cac HIS cgc ARG CDC cac HIS cgc ARG CDC cac HIS tac TYR CDC tcg SER tgg TRP CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC cac HIS tac TYR CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC cac HIS cgc ARG CDC cga ARG gga GLY CDC cac HIS tac TYR CDC tcg SER ttg LEU CDC cac HIS cgc ARG CDC tcg SER ttg LEU CDC cac HIS cgc ARG CDC cac HIS tac TYR CDC tcg SER ttg LEU CDC cac HIS cgc ARG CDC tcg SER ttg LEU CDC cac HIS ccc PRO CDC caa GLN gaa GLU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC cac HIS gac ASP CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC tcg SER ttg LEU CDC cac HIS ccc PRO CDC caa GLN cta LEU CDC caa GLN cta LEU CDC caa GLN gaa GLU CDC caa GLN gaa GLU CDC tcg SER ttg LEU 5

7 CDC cac HIS cgc ARG CDC caa GLN gaa GLU CDC cac HIS tac TYR CDC cac HIS gac ASP CDC tcg SER ttg LEU CDC cga ARG caa GLN CDC cac HIS cgc ARG CDC cac HIS cgc ARG CDC cac HIS cgc ARG 6

8 Table S2. Competitive fitness of in vitro selected rifampin-resistant mutants of M. tuberculosis. Strain rpob mutation Clinical frequency a Mean relative fitness (95% CI) Number of replicates Mutants in CDC1551 (Euro-American lineage b ) background: CDC01 S531L 54.0% 0.91 ( ) 6 CDC06 H526Y 11.0% 0.82 ( ) 7 CDC47 H526D 7.0% 0.78 ( ) 6 CDC07 S531W 4.0% 0.88 ( ) 6 CDC04 H526R 3.0% 0.82 ( ) 4 CDC10 S522L 1.0% 0.88 ( ) 6 CDC39 Q513L 0.7% 0.83 ( ) 4 CDC38 H526P 0.1% 0.84 ( ) 4 CDC49 R529Q 0.0% 0.58 ( ) 4 Mutants in T85 (East-Asian lineage b ) background: T8501 S531L 54.0% 0.96 ( ) 5 T8503 H526Y 11.0% 0.81 ( ) 4 T8538 H526D 7.0% 0.85 ( ) 4 T8505 S531W 4.0% 0.79 ( ) 4 a Data from 840 clinical rifampin-resistant strains summarized in ref. 29; b ref. 24; CI=confidence interval 7

9 Table S3. Rifampin-resistant mutants of strain T85 generated in vitro. Mutant rpob codon Wild type base pair Wild type amino acid Mutant base pair Mutant amino acid T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS tac TYR T tcg SER ttg LEU T tcg SER tgg TRP T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T cac HIS tac TYR T tcg SER ttg LEU T cac HIS tac TYR T cac HIS cgc ARG T tcg SER ttg LEU T cac HIS cgc ARG T tcg SER ttg LEU T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS tac TYR T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS cgc ARG T tcg SER ttg LEU T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS cgc ARG T cac HIS ctc LEU T tcg SER ttg LEU T cac HIS cgc ARG T tcg SER ttg LEU T tcg SER ttg LEU T cac HIS tac TYR T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS cgc ARG T tcg Ser ttg LEU T cac HIS cgc ARG T cac HIS gac ASP T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER tgg TRP 8

10 T cac HIS cgc ARG T tcg SER ttg LEU T cac HIS cgc ARG T cac HIS gac ASP T tcg SER ttg LEU T cac HIS gac ASP T tcg SER ttg LEU T cac HIS gac ASP T cac HIS cgc ARG T cac HIS cgc ARG T cac HIS gac ASP T cac HIS gac ASP T tcg SER ttg LEU T tcg SER ttg LEU T tcg SER tgg TRP T tcg SER tgg TRP T tcg SER ttg LEU T tcg SER ttg LEU 9

11 Table S4. Competitive fitness of clinically derived rifampin-resistant isolates of M. tuberculosis. Isolate pair M. tuberculosis lineage a Months Resistance 1 st isolate b between isolates c rpob mutation Mean relative fitness (95% CI) Number of replicates 1 Euro-American INH R 4 S531L 0.94 ( ) 4 2 Euro-American INH R 27 S531L 1.01 ( ) 4 3 Euro-American INH R 20 S531L 1.00 ( ) 4 4 East-Asian INH R + STR R nk S531L 1.01 ( ) 7 5 Euro-American INH R + STR R 18 S531L 1.16 ( ) 7 6 Euro-American susceptible 7 H526D 0.85 ( ) 4 7 Euro-American susceptible 6 H526D 0.71 ( ) 4 8 Euro-American INH R 18 S531W 0.82 ( ) 4 9 Euro-American INH R + STR R 4 H526R 0.82 ( ) 6 10 Euro-American INH R + STR R 6 H526Y 0.82 ( ) 4 a Ref. 24; INH R =isoniazid; STR R =streptomycin-resistant; CI=confidence interval; nk=not known. b antibiotic resistance profile of the first isolate; the second isolates had the same antibiotic profiles except for additional rifampin resistance; all resistance phenotypes were confirmed in the first and second isolates. c number of months between the rifampin-susceptible isolates and the rifampin-resistant isolate. 10

12 Table S5. Spoligotyping patterns of ten isolate pairs serially recovered from tuberculosis patients who acquired rifampin resistance during treatment. Pair Isolate Spoligotyping pattern 1 A B 2 A B 3 A B 4 A B 5 A B 6 A B 7 A B 8 A B 9 A B 10 A B 11